• Aucun résultat trouvé

Opportunities of the demographic dividend on poverty reduction in Sub-Saharan Africa

N/A
N/A
Protected

Academic year: 2021

Partager "Opportunities of the demographic dividend on poverty reduction in Sub-Saharan Africa"

Copied!
127
0
0

Texte intégral

(1)

This project is funded by the European Union under the 7th Research Framework Programme (theme SSH) Grant agreement nr 290752. The views expressed in this press release do not necessarily reflect the views of the European Commission.

Working Paper n° 47

Opportunities of the demographic dividend on poverty

reduction in Sub-Saharan Africa

UAM

Eva Medina and Sonia Chager

(2)

1 Framework Program 7

Grant Agreement no.: 290752 Project Acronym: NO-POOR

Enhancing Knowledge for Renewed Policies against Poverty

Instrument: Collaborative Project

Theme: SSH-2011.4.1-1: TACKLING POVERTY IN A DEVELOPMENT CONTEXT

DEMOGRAPHIC SCENARIOS AND THEIR IMPACT ON POVERTY:

Opportunities of the demographic dividend on poverty reduction in

Sub-Saharan Africa

WP9: Emerging issues & Poverty Scenarios in XXI Century Task 9.2.

February, 2016

Project Coordinator: Prof. Xavier OUDIN Work Package 9 Leader Organisation: E- UAM Research team: Dr. Sonia Chager, UAM Prof. Dr. Eva Medina, UAM

(3)

2

Contents

1. DEMOGRAPHIC WINDOW OF OPPORTUNITY AND DIVIDEND: DEMOGRAPHIC TRANSITION

AND ITS EFFECT ON THE ECONOMY ... 4

1.1. Relationship between demographic change and economic growth ... 4

1.2. Demographic Window of Opportunity and Dividend ... 6

1.3. When does the Demographic Window of Opportunity open? ... 9

1.4. Factors that affect the demographic dividend: policy to be implemented ... 14

2. POSITION OF SUB-SAHARAN AFRICAN COUNTRIES WITH REGARDS TO THEIR DEMOGRAPHIC WINDOW OF OPPORTUNITY AND DIVIDEND ... 18

2.1. Africa with respect to other regions ... 18

2.2. Hindering and promising factors to gain from a demographic dividend in Africa ... 38

2.2.1. Where are we now? ... 38

2.2.2. What could we do? ... 41

3. MODELLING THE DEMOGRAPHIC DIVIDEND AND ITS IMPACT THROUGH CONVENTIONAL ECONOMETRICS TECHNIQUES: THE CASE OF SUB-SAHARAN AFRICA ... 43

3.1. Literature review of empirical evidence on the demographic dividend ... 43

3.1.1. Recent empirical evidence on the demographic dividend in Sub-Saharan Africa ... 53

3.2. Methodology, variables and data: the construction of the panel database ... 56

3.3. Main results ... 61

3.3.1. Empirical results of the econometric specifications ... 61

3.3.2. Main policies for maximizing a Demographic Dividend in Africa ... 73

3.3.3. Ranking of African countries by key policies: who should focus on what? ... 79

4. SUMMARY AND CONCLUDING REMARKS ... 99

4.1. Objective, Methodology, and Data ... 99

4.2. Main Results ... 103

4.3. Policy Implications ... 110

5. REFERENCES ... 111

(4)

3

“The nature of the demographic transition and the way it has been experienced does much to explain just how human society got to where it is today. Differences in this process also help explain differences around the world today. Because the transition is a global phenomenon, the fact that various parts of the world are at different stages of the demographic transition helps us chart at least part of the future course among the relative newcomers to it. Forecasting the future is always risky and uncertain, but identifying the transition as a pathway to change enables us to understand more clearly the contexts of change to come for many countries” (Lee & Reher 2011, p. 2).

(5)

4 1. DEMOGRAPHIC WINDOW OF OPPORTUNITY AND DIVIDEND: DEMOGRAPHIC

TRANSITION AND ITS EFFECT ON THE ECONOMY

1.1. Relationship between demographic change and economic growth

Does population change have an effect on the economic growth of a country? Theories on this association date back to 1798, within a context of political debate, when Thomas Malthus stated that “the power of population is indefinitely greater than the power in the earth to produce subsistence for man”. Between 1798 and 1826 he published six editions of An Essay on the Principle of Population in which he pictured a negative scenario where excessive population growth would act as a strain to the limited supply of natural resources, which would in turn reduce the capacity for economic growth. Thus, in order to avoid such negative effects, Malthus asserted that there were two pathways to control the population: first, through what he called “positive checks”, that is mortality increase due to disease, famine, and war; and second, through “preventive checks”, which consisted of lowering fertility through moral restriction, including abstinence and delay of marriage, and also contraception, and abortion. Hence, Malthus and the subsequent Neo-Malthusians (for instance, Coale & Hoover 1958; and Ehrlich 1968) saw population growth as a restriction to economic growth. Moreover, the concern that high fertility causes poverty motivated being in favour of providing family planning programs in developing countries (Merrick 2002).

On the other hand, a different thought process emerged in the 60’s throughout the 90’s, where economists were beginning to reject the pessimist view as economic theory had begun to give increasing importance to technology and human capital accumulation (Bloom et al. 2003). Such “population optimists” viewed population growth as an economic asset. Kuznets (1960, 1967), Simon (1981), and also Boserup (1965, 1981) argued that with population increase, human ingenuity and resourcefulness also augmented. In other words, in times of adversity because of population growth, humans are stimulated to innovate (Boserup 1965, 1981) and, therefore, respond through the development of new technologies (progress on both agriculture and industry), as well as through social and institutional innovations (Sen 1999).

However, optimists took a broader view on the impact of population growth, suggesting a multiplicity of external factors that could have either positive or negative economic consequences (Bloom et al. 2003). As the authors point out (p. 16), “the broadening of the discussion on population growth eventually led to population neutralism emerging as the dominant view in the demographic debate”. Thus, taking into account that various studies have suggested that

(6)

5

population growth can be detrimental to economic development, or beneficial, depending on circumstances; the most recent view, “the neutralists” see rapid population growth as neither promoting nor impeding economic growth. The emergence of this view coincided with a declining interest in family planning as an instrument of economic development (Bloom et al. 2003).

In terms of the empirical analysis on this particular subject, even though earlier studies based on shorter time series found little statistical support for strong demographic effects (Kelley, 1988), supporting a rather neutral approach, the emergence in the 90’s of a theoretical framework by Robert J. Barro concluded that high fertility, population growth, and mortality all exert negative impacts on per capita output growth (Barro 1991, 1996; Barro & Lee 1994).

In turn, a new way of thinking (Chesnais, 1990) suggests that the debate may have been framed too narrowly (mostly focused on population size and growth rates), as little attention has been paid to a critical variable: the age structure of the population, which changes the way in which this issue is interpreted. In fact, taking the age structure into account provides powerful confirmation of the age-old view that, when it comes to the determination of living standards, population does, indeed, matter, as people's economic behaviour and needs vary at different stages of life, changes in a country's age structure can have significant effects on its economic performance (Bloom et al. 2003).

Precisely, the Harvard framework – consisting of several Harvard economists (e.g., Bloom & Williamson 1997, 1998; Bloom & Canning 2001, 2003; Radelet et al. 2001; Bloom et al. 2000) – emphasized the importance of including in the analysis the age structure concept. Indeed, a series of empirical studies based on aggregate level panel data concluded that demographic factors have a strong, statistically significant effect on aggregate saving rates (Schmidt & Kelley, 1996; Williamson & Higgins, 1997; Bloom, Canning & Graham, 2003; Kinugasa, 2004) and on economic growth (Kelley & Schmidt, 1995; Bloom & Williamson, 1998; Bloom & Canning, 2001). In fact, in a context of the economic miracle in countries from Eastern and South-Eastern Asia, detailed case studies thrived. The results of these analyses provided compelling and consistent evidence that the demographic changes were an important contributor to that region’s economic success (Bloom and Williamson, 1998; Mason et al. 1999; Mason, 2001, 2005).

All in all, the focus on “population impacts that take place due to imbalanced age-structure changes over the Demographic Transition” (in Kelley & Schmidt 2005, p. 276), indicated that population does matter to economic growth. The reasoning behind this analyses is that countries with higher proportions of people within the dependent ages (young and old groups) need to

(7)

6

spend more resources on them; while the opposite happens when having a higher share of working age population, who have more immediate returns (and can save and invest more) (Mitra & Nagarajan 2005).

1.2. Demographic Window of Opportunity and Dividend

The changes in the age structure of a population can have an impact on the economic performance of a country in the form of “demographic dividends or bonuses”. Such economic impacts occur as public expenditure (directed to social programmes - health, education) can be diverted towards investment in productive sectors and infrastructure. These benefits will decline once structural ageing starts to set in the society (increases in aged dependency burdens) (Pool, 2007). Hence, analysing the different periods of the demographic transition can help in understanding and identifying how and when a Demographic Window of Opportunity opens, in which a Dividend can be achieved (see Figure 1):

1) Pre-transition: high fluctuating death rates and equally high birth rates (Stage 1: low rates of natural increase). Before the demographic transition, population growth rates are relatively low owing to a combination of high mortality and fertility rates (CELADE, 2007), with mortality being especially acute among infants and children.

2) Early transition: high birth rates but diminished death rates (Stage 2: acceleration of natural increase). During this early transitional phase, population starts to grow rapidly as a consequence of a reduction in the mortality rates (mainly due to medical improvements, including health, hygiene, and sanitation, as well as improvements in food production, storage and transport), while fertility remains high. As a consequence of the mortality decline – especially infant and child mortality1 –, during this stage, the population structure becomes

increasingly youthful, where the size of “young dependents” is large, compared to the working group; hence, there is a high dependency ratio of those aged below 15 years old (Mitra & Nagarajan, 2005; Feyrer, 2007)

1 Precisely, the scene changes in the sense that it is not the kind of baby boom in which more babies are born, but

the one in which more babies survive and mature into children and adults (Bloom & Canning 2011). In fact, "Mortality declines were not evenly distributed across the population. Infectious diseases are particularly ruthless killers of the young, so their containment had the most powerful impact on the mortality of infants and children, which fell earlier and more quickly than mortality at other ages. The larger surviving youth cohorts served to drive down the average age of populations” (Bloom et al. 2003, p.26).

(8)

7

3) Late transition: declining birth rates (fertility transition2) and low death rates (Stage 3:

slowing natural increase). In the third phase, there is a reduction in the population growth rate as birth rates also start to decline3 while mortality remains low. It is within this stage that

there is a moment in which the last generation of infants born during the previous stage (a relatively large share) grow up, resulting in a bigger share of working age population (relatively high in comparison to both the young and old dependents) (Mitra & Nagarajan, 2005). The consequence is a rapid fall in the dependency ratio. Indeed, it is within this

particular transitional phase where the demographic window of opportunity opens, as the

proportion of working age population in the total population becomes the highest, with a potential growth inducing impact (Feyrer, 2007). In other words, it is where the working

population share peaks that the demographic dividend can be achieved (Pool, 2007).

4) Post-transition: Equally low death and birth rates (Stage 4). In this stage the population growth is negligible or even declines. Thus, the proportion of elder dependents rises relative to both young dependents and working age group and it is in this phase that the dependency ratio will be higher due to old age population (also called “population ageing”). Moreover, there has been increasing attention to the very low fertility observed in most post transitional societies (Bongaarts 2002). In this sense, the old aged dependency ratio can be even more accentuated in those more developed economies that register extremely low fertility levels, and that will have a greater reduction in the share of working age population in the near future.

2 See for instance: Van de Walle & Knodel, 1980; Caldwell & Caldwell, 1987; Caldwell, 1997; Dyson & Murphy,

1985; Castro, 1995; Rosero-Bixby, 1996; Cohen, 1998; Kirk & Pillet, 1998; Casterline, 2001; Heaton et al. 2002; Garenne & Joseph, 2002; Bongaarts, 2002, 2003; Lee 2003; Lloyd (ed.) 2005; Grant & Furstenberg, 2007; CEPAL, 2007; Shapiro & Gebreselassie, 2008; Machiyama, 2010; Garenne, 2011; Bongaarts & Casterline, 2013; among others. Moreover, other indicators such as Ideal Family Size (IFS) or the trends in unmet need for contraception can complement and explain the trends and pace of the fertility decline (Bongaarts & Casterline, 2013; Basu & Basu, 2014).

3There are plenty of explanations and factors behind the fertility decline, some of them being the higher

educational and employment opportunities for women (which can reduce the time available for childbearing), changes in nuptiality and union patterns (towards postponement), or even changes in the family size ideal as a consequence of lower infant mortality rates (use of contraception, abortion, family planning availability, or also pacing of births). In fact, demographers and social scientists have been engaged in active debate on the causes of low fertility and the prospects for further change (Chesnais 1996, 1998; Lesthaeghe 2001; Lesthaeghe & Willems 1999; McDonald 2000; Bongaarts 2002). Also, Becker (1960) advanced the argument that the decline in fertility was a by-product of the rise in income and the associated rise in the opportunity cost of raising children.

(9)

8

Figure 1: The Classic Stages of the Demographic Transition (Natural Increase of the Population)

Source: Population Reference Bureau (2004)

As seen above, the Demographic Window of Opportunity is initiated during Stage 3 of the Demographic Transition. It appears as the death rates tend to decline before the birth rates. Consequently, during this period in which mortality falls but not fertility (Stage 2) a baby boom is produced, which eventually ends when fertility rates decline. Economic growth is lowered with the baby boom (with the higher expenditure on child care), however, after 15-20 years, these large “boom” cohorts reach the prime years for working and saving, and it is within this moment when the country has an opportunity to rapidly grow and achieve a “Demographic Dividend”. Specifically, as pointed out by Bloom & Canning (2011), this impact can be generated through various pathways: 1) societies reallocation of resources from investment in children to physical capital, job training, technological progress and stronger institutions; 2) the rise of women’s participation in the workforce; 3) boost in savings.

Up to now the concept defined has been that of the first Demographic Dividend. Nonetheless, two phases of the demographic gift have been identified in the literature: a First Demographic Dividend, related to the population effect; and a Second Demographic Dividend, related to the productivity effect, which appears when the proportion of old age population increases, that is, the moment when the ageing of the population starts (Mason 2005, Mason & Lee 2006; Queiroz et al. 2006). In this second phase, wealth creation is relatively associated to population ageing, to the extent that

(10)

9

consumers and policy-makers are forward looking and respond to the demographic changes (Queiroz et al. 2006; Olaniyan et al. 2013). In other words, the prolongation of the retirement period, due to rising longevity, means that these economically dependent old aged individuals require one or more of the following: a) ownership of capital/savings accumulation; b) family support; and/or c) transfers from the state (welfare system). In any case, both dividends are related to one another and, as Mason (2005) asserts, the magnitude of the second dividend depends largely on how wealth is created during the first dividend.

1.3. When does the Demographic Window of Opportunity open?

A demographic window of opportunity is revealed as fertility rates decline, with a previous fall in the mortality rates, and the working age population increases relative to the dependent population. Thus, when identifying changes in the population age structure, in order to recognize the moment in which such a window of opportunity opens it is necessary to acknowledge various elements concerning a population’s birth, death and migration rates, by age and over time. These concepts are gathered in the basic demographic Balancing Equation of Population Change (Preston et al. 2001), which describes the only possible ways of entering or exiting a population:

𝑁(𝑇) = 𝑁 (0) + 𝐵 [0, 𝑇] − 𝐷 [0, 𝑇] + 𝐼 [0, 𝑇] − 𝑂 [0, 𝑇] ,

where N(T) and N(0) are the number of persons alive in the population at time T and at time 0, respectively; B [0, T] and D [0, T] are the number of births and deaths, respectively, in the population between time 0 and time T, in which the difference between the number of live births and the number of deaths in a time period gives the natural increase; I [0, T] is the number of in-migrations between time 0 and time T; and O [0, T] is the number of out-in-migrations from the population between time 0 and time T. Hence, this equation decomposes the changes in the size of the population into four flows, which can be translated into ‘rates’ if the size of these flows (number of occurrences) is related to the size of the population producing them.

In addition to these total rates, which reflect the “person-years lived in the population between time 0 and T”, it can be even more appropriate to account for age-specific variation when studying demographic events (see Preston et al. 2001 for more information). Namely, depending on the demographic outcome being studied, the age selection will be one or another: for example, infant mortality and adult mortality can have different patterns. Hence, selecting a suitable age group can

(11)

10

give different outcomes in terms of the opening of the Window of Opportunity, and the quantification of the potential Dividend.

Concretely, the opening of the window is related to the fall in the birth rates (preceded by decreasing mortality rates, specifically the infant mortality), while its ending is the result of having low fertility and mortality rates (both infant and adult rates). In other words, the demographic window appears at the beginning of the third stage of the Demographic Transition (late transition phase with declining birth rates and low death rates), that is, when a relatively large generation of baby boomers grow up and become potential producers, resulting in a bigger share of working age population. On the other hand, the window closes at the end of this third stage (where both birth and death rates become minimum), thus creating a change in the population pyramid or structure with an increase in the share of elderly individuals and decreasing young-adults, who will have to support the dependent population.

Nonetheless, this approach for measuring the timing of the Demographic Window can be somewhat limited in terms of designating the exact moment of closure, while the opening is more straightforward. As it can be seen in Figure 2, in the case of the Developed Regions, the patterns for Crude Birth and Death Rates nicely follow the theoretical stages of the Demographic Transition, showing the ending of Stage 3 in 1995-2000 approximately (with both rates becoming minimum), and the beginning of Stage 4 in 2000-2005 onwards. However, when it comes to the Less and Least Developed regions, the story is somewhat different. It appears that the Crude Death Rates start to increase before the Crude Birth Rates have reached their minimum (especially for the Less Developed). Additionally, the Crude Birth Rates in the Least Developed areas appear to not have yet attained their minimum. It seems that there will be a declining trend even beyond the 2100 scope. Hence, although the Demographic Transition stages offer an excellent approximation to the timing of the window, it is important to narrow down the time frame even more so as to gain in precision. Thus, several other methods to analyse the timing of the Demographic Window are presented next.

(12)

11

Figure 2: Demographic Transition by major economic regions4

Note: The solid line shows the Crude Birth Rates (average annual number of births per 1,000 population), while the dotted line shows the Crude Death Rates (average annual number of deaths per 1,000 population).

Source: Own calculations from the United Nations World Population Prospects (2012)

First and foremost, the window period can be further approximated with the use of proportions. The United Nations (2004) definition of the demographic window is the period during which the proportion of children and youth (younger than 15 years of age) falls below 30% and the proportion of elderly population (aged 65 and older) is still below 15%. Similarly, documenting changes in the Working Age Share (WAS) of a population is another suitable means to outline the beginning and ending of a positive demographic potential, through the growth and later fall in the proportion of working age people. In effect, the level at which the WAS peaks, and for how long in

4As defined by the United Nations World Population Prospects (2012):

(a) More developed regions comprise Europe, Northern America, Australia/New Zealand and Japan.

(b) Less developed regions comprise all regions of Africa, Asia (except Japan), Latin America and the Caribbean plus Melanesia, Micronesia and Polynesia.

(c) The least developed countries, as defined by the United Nations General Assembly in its resolutions (59/209, 59/210, 60/33, 62/97, 64/L.55, 67/L.43) included 49 countries in June 2013: 34 in Africa, 9 in Asia, 5 in Oceania and one in Latin America and the Caribbean. The group includes 49 countries - Afghanistan, Angola, Bangladesh, Benin, Bhutan, Burkina Faso, Burundi, Cambodia, Central African Republic, Chad, Comoros, Democratic Republic of the Congo, Djibouti, Equatorial Guinea, Eritrea, Ethiopia, Gambia, Guinea, Guinea-Bissau, Haiti, Kiribati, Lao People's Democratic Republic, Lesotho, Liberia, Madagascar, Malawi, Mali, Mauritania, Mozambique, Myanmar, Nepal, Niger, Rwanda, Samoa, São Tomé and Príncipe, Senegal, Sierra Leone, Solomon Islands, Somalia, South Sudan, Sudan, Timor-Leste, Togo, Tuvalu, Uganda, United Republic of Tanzania, Vanuatu, Yemen and Zambia. These countries are also included in the less developed regions.

(13)

12

time it remains relatively high, can capture the size of the window. Analogously, another ratio that provides similar information is the Ratio of Working Age to Non-Working Age Population (e.g. Bloom & Canning, 2011; Bloom et al. 2013; Olsen, 2012).

In conjunction to this train of thought, it is possible to obtain useful evidence on the window, seen as an approximation to the ratio of net consumers to net producers, by measuring the

Dependency Ratios5. Given that this ratio relates the population group most likely to be

economically dependent – that is, net consumers – to the group most likely to be economically productive, it provides an indication of the potential dependency burden that a country can face when advancing through its demographic transition. Precisely, the moment in which the dependency ratio falls (as fertility levels decline) and the proportion of working age population increases is when the demographic window of opportunity presents itself. It is within this phase that a society can make use of those savings – that would have otherwise been transferred towards large shares of dependent young population – towards productive investments or other social benefits. Besides, if the drop in the fertility levels persists and the share of potential producers decreases, that is, an ageing of a population comes about, the rise in the dependency ratio can indicate the end of said window. Thus, this calculation can not only give an idea of its total magnitude and timing but also if the numerator is disaggregated by age groups then it is also possible to differentiate between the population effect (decline in young dependents), that is the first dividend, and the productivity effect (population ageing), which would be the second dividend (e.g. Merrick, 2002; Mitra & Nagarajan, 2005; Bloom et al. 2007; Assaad & Roudi-Fahimi, 2007; Basu & Basu, 2014; Cleland 2012; Drummond & Thakoor, 2014). In a more detailed note, for example, the equations would be the following:

1) Total Dependency ratio = 100 x (𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑎𝑔𝑒𝑑 0−14) + (𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑎𝑔𝑒𝑑 65 𝑎𝑛𝑑 𝑚𝑜𝑟𝑒)(𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑎𝑔𝑒𝑑 15−64) ;

2) Child/Youth Dependency ratio = 100 x (𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑎𝑔𝑒𝑑 15−64)(𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑎𝑔𝑒𝑑 0−14) ;

3) Old Age Dependency ratio = 100 x (𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑎𝑔𝑒𝑑 65 𝑎𝑛𝑑 𝑚𝑜𝑟𝑒)(𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑎𝑔𝑒𝑑 15−64)

5 The methodological approach for the dependency ratio and its definition are extracted from the United Nations

World Population Prospects (2012 Revision), and the United Nations methodology sheets: http://www.un.org/esa/sustdev/natlinfo/indicators/methodology_sheets/demographics/dependency_ratio.pdf

(14)

13

In general, the age selections are often established at 15 to 64 for the working age population, where the youth dependency ratio would be that of those below the age of 15, while the elderly dependents are habitually aged 65+. Yet in some cases the age classifications are somewhat different, such as in Mitra & Nagarajan (2005), who use the elderly group 60+, or Basu & Basu (2014) who consider the ages 20-64 as the working age population in order to account for increasing years of education. Other authors do, however, disaggregate even more so the age groups, especially when using a cohort approach with narrower age-ranges and more refined data, which allows to observe the degree of turbulence that takes place from population waves and depressions as cohorts of varying sizes ‘flow’ through key life cycle stages (see Pool, 2007). Along these lines, it is worth mentioning another more elaborate analytical approach in the literature that computes Support Ratios, also in terms of producers to consumers, that incorporates a weight in its calculation so as to allow for age-specific variations (e.g. Mason & Lee, 2006; Fürnkranz-Prskawetz et al. 2013). Indeed, this particular method is a much more nuanced approach, as it takes into consideration the variation in productivity at different ages as well as differences in consumption needs (Oosthuizen, 2015). The main source of information, in this case, often comes from consumption and labour income data and, more concretely, its age profiles,

made available by the National Transfer Accounts (NTA) framework6. In order to construct such

estimates, the NTA focuses on what is referred to as the generational economy, which is defined as: ‘‘1) the social institutions and economic mechanisms used by each generation or age group to produce, consume, share, and save resources; 2) the economic flows across generations or age groups that characterize the generational economy; 3) explicit and implicit contracts that govern intergenerational flows; 4) the intergenerational distribution of income or consumption that results from the foregoing’’ (Lee & Mason, 2011: p.7). In effect, the disaggregation is carried out for each year of life, adding a weight of the ages according to the consumption and income associated to each year of life. Regarding the timing of the window, the onset is also often determined by the period in which the support ratios fall or when the share of working-age people increases; while the end date is given by increasing dependency ratios and decreases in the proportion of potential producers. Likewise, peak moments can be identified, with support ratios becoming minimum for instance. All in all, even though this latter specific method overcomes the age concern/limitation when using the dependency ratios, as it has no explicit (arbitrary) age cut-offs, for the purpose of the present study, the number of countries with NTA data are not sufficient. Moreover, regarding

6 See the National Transfer Accounts Database (NTA), which provides consumption and labor income profiles for

(15)

14

the NTA profiles for African countries, only 7 countries are available at present: Benin, Ghana, Kenya, Mozambique, Nigeria, Senegal, and South Africa.

1.4. Factors that affect the demographic dividend: policy to be implemented

The opening of a demographic window of opportunity does not always guarantee a surge in economic growth. Indeed, collecting the dividend is not automatic as the intensity of the age-structure effect depends on firstly, the speed and magnitude of the demographic transition and, secondly, on economic and social conditions favouring to exploit this gift (See Figure 3).

Primarily, concerning demographic factors, the change and pace of the demographic transition depends on the timing and intensity of both the mortality and fertility declines, as well as the population inputs and outputs from the countries migration history and present. Secondly, with regards to economic and social conditions, various issues have been identified in the literature as advantageous for obtaining a better utilization of the Demographic Window of Opportunity: institutional setting and political stability7 (Bloom and Canning 2001; Mitra & Nagarajan 2005;

Bloom et al. 2007); employment levels (Mitra & Nagarajan 2005; Cleland 2012; Olsen 2012; Fürnkranz-Prskawetz et al. 2013); the rise of women’s participation in the workforce (Merrick 2002; Bloom & Canning 2011); education and human capital (Assaad & Roudi-Fahimi 2007; Lutz et al. 2008; Cuaresma et al. 2014; Drummond et al. 2014); etc.

It is possible to recollect examples of countries experiencing different trends with regards to these factors. For instance, concerning the decline in birth rates, the transition from high levels (i.e., five to eight children per woman) to low levels of fertility (i.e., two or fewer children per woman) can span approximately 60 years (or 2 generations) – being considered as a smooth transition –, or it can occasionally occur much faster (over 15-25 years, e.g., some Asian countries), or slower (over a century or more, e.g., most of Europe), or even stall (e.g., Argentina) (Garenne 2011). On the other hand, regarding the economic and social conditions, good governance and judicious public policies are required, as evidenced by Latin America when it experienced demographic conditions similar to East Asia after 1970, but GDP per capita grew by only 0.7% annually over 1975-1995, compared to 6.8% in East Asia (World Bank, 2008; Bloom et al. 2010). A combination of rigid labour markets, a relatively closed economy, an inability to attract investment on a sufficient scale, and weak governance led to a period of stalled growth in Latin America (Bloom et al. 2010;

7 In this sense, an unfavorable setting can explain the failure of Latin America and the North African countries to

(16)

15

Drummond et al. 2014). In Asia, Merrick (2002) gives examples of successful countries such as South Korea and Taiwan; and policy failures in India and Bangladesh (who might not benefit from the favourable demographic conditions).

All in all, in order to take advantage of the Demographic Window of Opportunity and increment the magnitude of the Demographic Dividend demands the concern from governments and policy makers oriented towards areas such as family planning, health, education, gender equality and employment generation and political stability. In particular, those actions that permit an increase in the capitalization of the Demographic Dividend can be classified8 as follows:

Demographic:

 Lower infant and child mortality: through the expansion of childhood immunization and

the provision of safe water and sanitation; access to health services and medical care; reduction in levels of malnutrition; increase in mother’s education by enhancing their health knowledge; reduction of adolescent pregnancies; limit selective abortions and fight against later neglect on daughters; access to information on HIV/AIDS and health provision.

 Encouraging fertility reduction9: access to primary and reproductive health services and

family planning programs, and to girls’ education; reducing unwanted pregnancies through better information and access to contraception; delay in the age at first union/marriage; reduction in the prevalence of early and child marriage; entry of women in the labour force; gender equality.

 Tackle adult mortality: for example, in the case of high prevalence of HIV infection, which affects primarily the young-adult population, in order to avoid an increase in orphaned children and a lack of support for the elderly (in a context of absence of social welfare).  Knowledge of migration flows: identify poles of attraction and push-pull factors.

8 The primary source for this section is Bloom, Canning and Malaney (2000); Merrick (2002); Bloom, Canning and

Sevilla (2003); Ross (2004); Bloom et al. (2007); Lutz et al. (2008); Kautz et al. (2010); Bloom & Canning (2011); Olsen (2012); Bloom et al. (2013); Bongaarts and Casterline (2013); and Juhn et al. (2013).

9 As Merrick (2002) states: there is a belief that economic policies determine poverty reduction and that

contraception is a "private good", where economists argued that the correlation did not necessarily imply causality, in fact, the relationship could run the opposite direction - poverty could be the cause of high fertility (poor people often want more children because offspring represent wealth, provide household labour and are the only form of social security for ageing parents).

(17)

16

Economic and Human Capital:

 Good governance that provides political stability (helping pursue policies with continuity and success), institutional quality and efficient infrastructure.

 Improve human resource capabilities: invest in human capital by providing universal primary education, as well as secondary and higher education so as to improve the mismatch between skills and labour market needs.

 Job creation is essential: permitting to absorb large numbers of potential workers in the

labour market.

 Open trade policies with open economies: in order to obtain a faster growth during the limited time afforded by the window.

 Policies to generate capital (encourage savings, on the personal level, but also by the government and businesses), and attract foreign investments and development assistance (efficient financial and labour markets).

 Improve gender equality: increase female education and labour force participation, as

more education for women can be translated into more women in the paid labour force.

Overall, the demographic transitions acts as a window of opportunity to earn a demographic dividend (Carvalho and Wong, 1999; Pool, 2007), when it is combined with good policies (Bloom and Canning, 2001). Thus, Asia’s more favourable outcomes have been attributed to a stronger focus on human (education and health) and physical capital as increased employment opportunities and higher labour participation rates, including for women, allowed Asia to maximize the benefits from the increase in labour force; while, on the other hand, a weak policy environment hindered Latin America’s ability to benefit from a demographic dividend

(18)

17

Figure 3: Demographic Transition, Window of Opportunity and Dividend

(19)

18 2. POSITION OF SUB-SAHARAN AFRICAN COUNTRIES WITH REGARDS TO THEIR

DEMOGRAPHIC WINDOW OF OPPORTUNITY AND DIVIDEND

2.1. Africa with respect to other regions

The demographic transition has swept the world since the end of the nineteenth century (Galor 2012), starting in many European countries and parts of the Americas and is currently underway in most of the world, and the presumption is that it will eventually affect all countries (Lee & Reher 2011). As the latter authors suggest, no country has fully completed this process, since mortality decline will most likely continue, and population aging lies mainly in the future. Nonetheless, concerning the total fertility rates, the majority of countries that have experienced large and rapid falls levels towards fertility levels below 2 births per woman are in the more developed countries (see Figure 4, first graph), while those nations who are at the opposite side of the spectrum, the least developed countries, reveal a later fertility decline starting from much higher levels (above 6 births per woman) and an expected longer period of reduction over time.

In fact, if one takes a look at the fertility decline by geographical areas (Figure 4, second graph), with regards to the developing regions, it began in the mid-1960s in Latin America and slightly later in Asia, the reductions being for both steady, rapid, and with similar paths. In particular, East Asia has had the fastest and most pronounced demographic transition in history (Bloom et al. 2013). On the other hand there is Africa, and more concretely Sub-Saharan Africa (SSA), which is still the region that stands as an “outlier” with high fertility rates that are slowly falling. Precisely, the fertility transition in Sub-Saharan Africa did not begin until the late 1980s with a fertility rate in the period 2005–2010 of 5.1 children per woman, which was still more than double the level observed in the other two regions (Boongarts & Casterline 2013).

(20)

19

Figure 4: Total Fertility a by major areas and main geographical regions, for the period 1950-2100, expressed

as children per woman (Medium Fertility Scenario)

a: Average number of children a hypothetical cohort of women would have at the end of their reproductive period if they were subject during their whole lives to the fertility rates of a given period and if they were not subject to mortality

(21)

20

Additionally, the speed of the transition is also lower in Africa than in other regions, mainly due to factors such as child mortality, HIV/AIDS and general health, sanitation and immunization problems, as well as political instability, wars and high net emigration (Olsen 2012). Still, there is significant heterogeneity across the African countries in terms of when they started the transition: South Africa, Botswana, Cape Verde, Seychelles and Mauritius have nearly completed the demographic transition, in a time frame similar to that of Asia and Latin America (Drummond et al. 2014). More so, heterogeneity within countries also exists which can be reflected through differences between urban and rural areas. In this regard, Garenne & Joseph (2002) found that, in most of Africa, the common pattern that emerged from their analysis was an early fertility decline in urban areas followed about a decade later by the rural areas. As Eastwood & Lipton (2011, p. 13) assert, each demographic transition phase “arrives earlier, and proceeds faster, for urban, richer, more economically integrated, and less gender-unequal groups and places”. Thus, national trends in total fertility or the pace of natural increase can be somewhat misleading, if one does not take into account the diversity within the region.

All in all, not only is Africa behind other regions in terms of the fertility transition, but the continent also suffers from other issues regarding the mortality transition. In this sense, the reality of HIV/AIDS can thwart the advances and gains achieved in terms of life expectancy and decrease in the overall mortality rates, which can also in turn affect the fertility rates10. Precisely, the fact

that the demographic transition is yet to take place has been provided as one of the reasons holding back growth in the region (Bloom et al. 1998; Bloom et al. 2003). However, as Ross (2004, p.4) bestows, Sub-Saharan Africa is just starting to enter its demographic window of opportunity – under the assumption of declining rates over the next several decades11 – and if these declines

persist and if governments involved take pro-active actions following to some extent those as from East-Asia, “the dividend may become real rather than potential”.

10If a country suffers a high prevalence of HIV infection, the simple physiological aspect of the disease can also

affect fertility, as infected women are significantly less likely to give births than non-infected (Juhn et al. 2013).

11In the case of Africa, the decline seems to be smooth and continuous, yet some authors argue (see bibliography of

Garenne 2011, list from 5 to 16) that in some countries fertility has stalled (yet there is a lot of debate). Fertility stalls appear uncommon in African countries (of the 31 countries investigated in earlier studies, only 8 - Ghana, Kenya, Madagascar, Nigeria, Rwanda, Senegal, Tanzania and Zambia - exhibited some kind of stall, and mostly of short duration - less than 10 years) (Garenne 2011).

(22)

21

In fact, according to the United Nations World Population Prospects (2012), the fertility transition in SSA has already begun as the Crude Birth Rates (average annual number of births per 1,000 population) have been following a declining trend from the period 1985-1990 onwards, parting from 47.4 births per 1,000 population in the 50’s to 44.8 in the late 80’s, with the latest estimate being 39.5 in 2005-2010 (See Figure 5). In effect, even though in all three fertility scenarios the pattern is towards a descent in the birth rates, in a context of low fertility the drop would be much more pronounced and rapid, whereas the opposite would happen in the high fertility scenario. Therefore, in terms of the Demographic Transition, it is possible to acknowledge that Sub-Saharan Africa timidly started its third stage of the Demographic Transition during the late 80’s and beginning of the 90’s, as the crude birth rates initiated a slow but steady decline. Additionally, except for the slight stagnation during the decade 1990-2000, the crude mortality rates have been following a solid decrease that is forecasted to come to a halt in the 2040-2045 period (remaining rather stable in the high fertility scenario, and gradual and more pronounced increases in the medium and low fertility scenarios, respectively, mostly due to the ageing process that would come about). It is important to notice that when identifying the window with the demographic components of natural increase, it is possible to outline the beginning of said window, and to a certain extent its completion with the mortality rates. However, the United Nations (2012) forecasts for the birth rates are that of ongoing declines, even beyond the scope of 2100. Therefore, if we consider the ending of the window as the moment in which the birth rates become minimum, in the case of Sub-Saharan Africa we will still have to wait for newer predictions that estimate the lowest fertility rates achieved in the future.

(23)

22

Figure 5: Demographic Transition in Sub-Saharan Africa

Source: Own calculations from the United Nations World Population Prospects (2012)

Hence, for any country of interest, pinpointing its demographic transition stages, especially for those who have already entered this third stage becomes essential, with its demographic components12 of natural increase being the ‘spoiler’ of de window period. Therefore, the next step

is to narrow down the opening and closure of the window by applying various alternative methods in order to provide a more accurate timing with the data available. The first option is to employ the United Nations definition of the Demographic Window of Opportunity (United Nations, 2004), which delineates the window as the period in which the proportion of children (aged less than 15 years old) is less than 30%, while the proportion of old aged (65+) is below 15%. The results are observed in Figure 6:

12 It is also worth mentioning that, as Olsen (2012) points out, there is a lack in the literature when it comes to the

component of migration, and even more so when taking into account the African context: “Most empirical studies

analyzing the demographic dividend are based on data aggregated to country level. They have largely ignored migration and the role of regional labor market integration, even though increasing youth unemployment is leading to emigration of young Africans in the search for job opportunities beyond national borders” (p.3).

(24)

23

 Europe entered its demographic window before 1950 and has recently finished it around the

years 2000-2004 where the proportion of old aged was 14.7%, with Northern Europe having trespassed the 15% threshold a decade earlier. Since then, the pattern has been towards a significant ageing of the population, not only because of the relative stability in the proportion of children around 16% (2005-2100) but due to the relentless increase in the forecasted proportion of old aged over time, especially until 2055 (reaching a proportion of 28%). Northern America, initiated its window in 1970 and will be closing it in 2015, while Oceania’s window is shifted towards a decade later start. Indeed, compared to Europe, both Northern America and Oceania are projected to have a slightly younger populations.

 On the other hand, Asia and Latin America follow a similar pattern starting the window around 2005 and ending it in 2035 and 2040 for Latin America and Asia, respectively. It is rather important to note that the ageing process in Latin America is projected to be the highest for all regions. Furthermore, Eastern Asia appears to be a region with a tremendously rapid decline in the proportion of children over time, having the earliest onset for the whole Asian region around 1985 (see Annex, Figure 6.2) and a striking increase in the proportion of old aged, finishing the window around 2025, and a noteworthy ageing of the population in the near future.

 Finally, concerning Africa and Sub-Saharan Africa specifically, the pattern is rather distinctive with respect to the other major regions. Africa as a unity is expected to initiate its window of opportunity much later in time, around 2060, with Sub-Saharan Africa starting in 2065. As seen previously with the demographic transition phases (components of natural increase, especially fertility rates), it is still not possible to delineate the completion of the window. In fact, using the United Nations delimitations, Africa and Sub-Saharan Africa will probably surpass the 15% of old aged beyond the year 2100. However, the continent faces notable heterogeneity when one takes a glance at the timing by regions. Southern and Northern Africa exhibit a similar pattern to that of Asia and Latin America, yet marginally delayed in time. Southern Africa is about to open its window of opportunity, in 2015, and close it around 2070, while Northern Africa’s window is expected to be somewhat shorter in time, between the years 2025-2060. Of the remaining regions, Eastern Africa is expected to be the next in line to open and finish its window (2060-2095). Finally, although Middle Africa and Western Africa are projected to be the last ones to access the demographic opportunity, in 2065 and 2075 respectively, they are also the two regions with a question mark with regards to the closing date.

(25)

24

Figure 6: Demographic Window of Opportunity13 (United Nations definition)

Source: Own calculations from the United Nations World Population Prospects (2012)

13 The solid line shows the proportion of children below the age of 15, while the dotted line shows the proportion

(26)

25

So far, this method has accomplished the task of identifying the timing, and to a certain extent the size of the demographic window of opportunity. Analogously, if one takes a look at the actual share of working age population, documenting its changes in the pattern over time can also shed light to its timing, and especially the magnitude of the window. In Figure 7 the variation in the Working Age Share (WAS) by major regions and for the period 1950-2100 is presented (for the United Nations Medium Fertility Scenario). The comparison among regions shows Eastern Asia as the positive outlier, who is at the moment experiencing its highest peak in its proportion of young-adult population, with a WAS of 72.6% in 2010 (following a trend of more than 70% of WAS already in 2005, which is projected to continue until 2015). In fact, it is the region with the maximum share of working aged people achieved, and projected to be reached. Subsequently, the tallest peaks are also within the same time period 2005-2010 for Asia (as a whole), and Europe, North America and Oceania, with shares close to 68% (except Oceania with 65%). Similarly, the rest of Asia is also projected to accomplish their highest WAS peaks around 68%, with South-Eastern Asia realizing it around 2025, Western Asia in 2035, and Central and South-Central Asia in 2040. However, Latin America and the Caribbean is expected to attain slightly lower sizes in its window, with peaks ranging from 66% in the Caribbean and Central America (in 2020 and 2030, respectively) to 67.5% in South America in 2020. Furthermore, the African continent is also forecasted to achieve lower peaks, in this case ranging from 68.4% in Southern Africa (around 2045), 66.3% in Northern Africa (2040), 65.8% in both Middle and Western Africa (2085 and 2095, respectively), and finally 64.1% in Eastern Africa (around 2070).

(27)

26

Figure 7: Working Age Share around the globe

Source: Own calculations from the United Nations World Population Prospects (2012)

Nevertheless, these peaks could become somewhat higher if we took as reference the Low Fertility Scenarios from the United Nations projections (see Annex, Figure 7.2), especially for Africa. In fact, the window would start earlier for Sub-Saharan Africa in a low fertility scenario, having at the same time a relatively larger WAS when compared to the other two fertility scenarios, the medium and mainly the high fertility setting. In this sense, if we arbitrarily chose a threshold of 60% of the share in the working age population, the window would open around 2030 in the low fertility scenario, which would be 15 and 30 years earlier than in a medium and high fertility situation, respectively. Concerning Asia and Latin America, where it is possible to observe the opening and closure of a demographic opportunity, measuring the proportion of young-adults who can be potential workers can help sketching the widening and narrowing of the demographic window of opportunity. For instance, the widening of Asia’s window is expected to be rather sizeable over time, of approximately 95 years if we consider a WAS of over 60% (although it would close much quicker in the low fertility setting). Eastern Asia and Latin America and the Caribbean, however, would appear to have a narrower windows of 70 and 65 years, respectively, with a WAS above 60% in the medium fertility scenario. Nonetheless, it is worth mentioning that both these regions exhibit less variation the ending of the window when comparing between fertility scenarios.

(28)

27

Thus, having presented the Demographic Transition stages, as well as the proportions of children and old aged, and the share of working age population as indicators of the timing and size of the demographic window of opportunity, the following step is to relate the population that is a net consumer to the group most likely to be economically productive (net producer).Up to this point, the calculations have had a common denominator: the total population. However, in order to approximate even more the moment in which the dependency burden becomes minimum, the Dependency Ratio is an appropriate method that can be employed to measure it, as it indicates the potential effects of changes in population age structures by drawing attention to the broad trends in social support needs. Hence, in this case, the denominator used is different, that being the population aged 15-64. Additionally, when the numerator is disaggregated by age groups it is possible to distinguish between the population effect (decline in the young dependents), and the productivity effect or population ageing (increase in the old aged dependents), which would equate to the first and second dividends literature.

Therefore, Figure 8 displays the Total, Child and Old Aged Dependency Ratios by major regions and for the three fertility scenarios. The values of these ratios are understood in this manner: for every 100 working age population you have x number of dependent people (since the unit of measurement is per hundred persons aged 15-64). Here the demographic window of opportunity is established through the combination of both child and old age support burden where the idea is to accomplish a minimum Total Dependency Ratio (TDR). Therefore, the opening of the window is provided by the decline in the child ratio (as the fertility transition takes place), while it’s shut down is given by the rise in the old aged burden ratio (once the society reaches low levels of fertility and mortality). Hence, the window can be defined as the period with a decline that comes about right after the highest peak in the TDR until it becomes minimum or starts increasing again. The results show the following:

 Throughout the 50’s-60’s, Europe, North America and Oceania set off with the lowest levels of

Total Dependency Ratios (around 55-68), while on the other end of the spectrum, there was Central America (96), Eastern Africa and Northern Africa (almost 90).

 The first regions to initiate the descent in their TDR (the following period right after their highest peak) were Europe (who already began with low levels), North America and Oceania in 1965; then in 1970 it was the turn for Latin America and the Caribbean and most of Asia (except Central and South-Eastern Asia, who began 5 years later); and finally, the African regions who exhibited differing trends, in which some followed similar patterns to that of Asia

(29)

28

and Latin America such as Southern Africa (1970) and Northern Africa (1975), while Western and Eastern Africa would have to wait until 1995, and Middle Africa until the year 2000.  In terms of the speed of decline, Eastern Asia has had one of the fastest declines worldwide in

their Total Dependency Ratios. The region dropped from a ratio of 76.4 in 1965 (its peak) to 37.7 in 2010 (lowest point), achieving a rate of change of -50.6% in 45 years. In this case, the transformation has taken place within the “estimates” period, that is, Eastern Asia’s TDR reached its lowest point in 2010, whereas in all three UN fertility scenarios, the trend is towards an increase of their dependency burden. In effect, it has been the only region from the developing world to have achieved such a feat. On the other hand, regarding some of the other regions, within the Medium Fertility Scenario, the speeds on their dependency shifts are as follows: Central America, for instance, is expected to also develop an intense alteration in its population structure, moving from its highest ratio in 1965 (100.7) to its lowest in 2030 (51.0), which would be a rate of change of -49.3 in 65 years; South America, nonetheless, is expected to experience a similar outcome, although relatively faster, with a rate of change of -43.85 in 55 years (from a ratio of 85.5 in 1965 to 48.0 in 2020); finally, with respect to the two regions in Africa who are more advanced in the demographic transition process, Southern and Northern Africa are forecasted to drop their TDR with a similar rate of change of 45.6 and 45.9, respectively (with a TDR of 85.3 in 1965 to 46.1 in 2045, that is 80 years for Southern Africa, while for Northern Africa the change would be in 70 years, from a ratio of 93.5 in 1970 to 50.82 in 2040).

 Precisely, in the estimates section, that is between the years 1950-2010, and especially during the period 1950-1990, while all of the regions had been experiencing declining trends in their respective TDRs, the ratio for the Sub-Saharan region and more specifically, the Western, Eastern and Middle areas, was becoming higher (reaching ratios higher than 90). In fact, these regions are also predicted to be clear outliers in the three fertility scenarios, displaying an opposite trend: when the rest of the world is facing an increase in their dependency burden, these areas are expected to be decreasing it. This same pattern holds true also for the Child Dependency Ratio (CDR) and the Old Aged Dependency Ratio (ODR). Moreover, the graphs illustrate great differences in the dependency levels of these regions compared to the others, which can be clearly seen for the ageing dependency, but remarkably so for the child dependency pattern over time, and for the three fertility scenarios. These disparities are a strong indication that most of Sub-Saharan Africa is still far behind in its demographic transition process. Certainly, a good example is Western Africa as it emerges as the area in

(30)

29

which children still dominate its age structure, thus, having a lower projection of older dependents over time. When observing the fertility scenarios for the child dependency ratio, the region will likely catch up with the rest of the world in the distant future (well beyond 2100 and even more so in the low fertility scenario). However, although the forecasted old age dependency ratio is expected to start rising well after 2050 (in the medium fertility scenario) and even later in a high fertility setting (around 2070), it is very unlikely that it will reach ratios similar to those worldwide given its great distance.

(31)

30

Figure 8: Total, Child, and Old Age Dependency Ratios by Major regions, for all three Fertility Scenarios

(32)

31 Source: Own calculations from the United Nations World Population Prospects (2012)

(33)

32

There are various ways to interpret when a window of opportunity will present itself. Until now, we established it as the period right after the highest to the lowest peak, that is, when the age structural shift is taking place. However, as observed before, not all regions depart from the same levels. In addition, it is also rather important to take into account the differences in levels depending on the UN fertility scenarios, especially in settings where the fertility transition is still in process, and even more so in those contexts in which the fertility rates are still very high, as would be the case for Sub-Saharan Africa, as it is a region where the Total Fertility Rate in its latest estimate (2010-2015) is still of 5.1 children per woman. Therefore, in figure 9 the Total, Child, and Old Age Dependency Ratios for Sub-Saharan Africa, Asia, Eastern Asia, and Latin America and the Caribbean (for the three fertility scenarios) are illustrated.

First, concerning Asia, the region attained its highest TDR in 1965, which was mostly due to a high child dependency burden. Precisely, in the time frame where the child ratio was heavily declining, the old age burden was slightly increasing. In terms of the Fertility Scenarios, the window is projected to close around 2010 (Low), 5 years later (Medium), and around 2025 (High). More so, despite the child ratio being projected to remain relatively stable over time for all scenarios, the Old Aged Dependency Ratio is another story, which can become especially troublesome in the low Fertility Scenario. On the other hand, Eastern Asia’s age structural transformation, as mentioned previously, took place within the time frame 1970-2010, and analogously to what is expected to occur in Asia with regards to the increase in the old age burden, for Eastern Asia the Old Aged Dependency Ratio is forecasted to exceed a value of 70 (in the low scenario), which is higher than that for Asia as a whole (ratio of 60).Similarly, the demographic window for Latin America and the Caribbean is akin to that for Asia. Latin America’s peak in the TDR was also approximately in 1965, where the most important contribution to the drop in the ratio was given by the decrease in the child burden, which will end around 2015 (High Fertility), 2020 (Medium Fertility), and 2025 (Low Fertility). However, for Latin America the Old Aged Dependency ratio is expected to be even more eminent (ratio of almost 90 in 2100).

Secondly, on the other side of the spectrum we have Sub-Saharan Africa. The demographic opening, provided by the fall in the TDR was initiated around 1990, however it sets off from really high levels, with a ratio of 93.5 dependents for every 100 working age people, where the Child Dependency Ratio at that time was already 87.8. Hence, the Old Aged Dependency Ratio was at its minimum (5.7). Unlike the other regions, in SSA while the child burden is expected to substantially drop, the old age burden is forecasted to remain pretty much stable for most of that time, only beginning to slightly increase at later stages (around 2045-2050). In effect, it is in this region that the differences between fertility scenarios become acute: the window is projected to close around

(34)

33

2070 (Low), 2085 (Medium), and 2095 (High). Thus, if the high trend in the Total Fertility Rate continues to prevail in much of SSA, the demographic window of opportunity will not only close much later in time, but its minimum TDR will also be higher (ratio of 59 in the high setting as opposed to a ratio of 50 in the low fertility situation).

Figure 9: Demographic Window of Opportunity depending on the Total Dependency Ratio for Sub-Saharan Africa, Asia (and Eastern Asia), and Latin America and the Caribbean.

(35)

34 Source: Own calculations from the United Nations World Population Prospects (2012)

(36)

35

Nevertheless, one cannot omit the reality that in Africa there is a great deal of heterogeneity. As it can be observed in Figure 10, that illustrates the Demographic Window of Opportunity depending on the Total Fertility Rates (for the Medium Fertility Scenario), again expressed as the period right after its highest peak to its lowest one, by major regions, and for individual African countries. Concerning the developed countries, that is Europe, Northern America, and Oceania, these are the regions in which the demographic window has opened the earliest, around 1965, followed by Latin America and the Caribbean in the 1975, and Asia in the 80’s. Once more, Eastern Asia stands out as the region in which the transition of the dependency ratio from high to low has been the fastest and most intense (as can be seen in the changes in colour at every consecutive period towards bluer tones, that is TDRs below 40). The rest of Asia will apparently take relatively longer and will achieve values around 50-55. On this regard, theoretically speaking, the pace and speed of the fertility decline affect the duration of the demographic window of opportunity, including its level (TDR) at the end of the window. As Birdsall et al. 2001 state, the potential benefits from having a low total dependency ratio become larger when the decline of the demographic transition is fast, however, the period of the window will be shorter. In other words, as it can be observed in Figure 10, Eastern Asia’s window is the shortest yet the structural transformation from high to low TDR is outstanding, reaching the bluest tones at the end of its window. In fact, another explanation can be found in the fact that the total dependency ratio takes into account both dependent populations, that is, the children and elderly. In this sense, as it could be seen in Figure 8, the TDR is affected first by the decline in the child dependency ratio, and much later in time, as the old age dependency ratio increases, the decreasing trend of the TDR slows down and begins to increase. Hence, if the speed of the demographic transition is fast, the window is shorter, which means that by the time the window ends, the ageing of the population has barely affected the TDR. However, if the pace of the transition is slow, instead of reaching blue tones, these windows will end up having greener ones since the increase in the old age dependency ratio in the later years will also increase the TDR.

Regarding the African countries specifically, although the demographic window for Africa, as a whole, begins around the year 1990 (and five years later for Sub-Saharan Africa), there are a few individual countries that open their window much earlier, yet with higher levels in their respective TDRs (more orange tones). These are the following: Mauritius (1965); Reunion, Algeria, Tunisia, Botswana, and South Africa (1970); Seychelles and Morocco (1975); Djibouti, Mayotte, Libya, Cape Verde (1980); and Zimbabwe, Sudan, and Ghana (1985). Additionally, some African countries exhibit striking analogous patterns as that of Eastern Asia, in terms of the speed of the transition: Mauritius, Seychelles, Reunion, and to a lesser extent Zimbabwe (in Eastern Africa); Libya, Morocco, Tunisia, Western Sahara (in Northern Africa); and Cape Verde (in Western Africa); and a

Figure

Figure 1: The Classic Stages of the Demographic Transition (Natural Increase of the Population)
Figure 2: Demographic Transition by major economic regions 4
Figure 3: Demographic Transition, Window of Opportunity and Dividend
Figure 4: Total Fertility  a  by major areas and main geographical regions, for the period 1950-2100, expressed
+7

Références

Documents relatifs

In the larger group of sensorimo- tor DPN, distal symmetrical polyneuropathy (DSP) is the most common type of diabetic neu- ropathy [13]. Patients suffer pain, sensory

This research follows on from a previous study we did on the excess mortality of twins in SSA, using national surveys (Ouedraogo et al., 2021).The issue was that of excess

The LIDE variable measuring foreign direct investment has a negative impact on the Gini index to a threshold of 5%.For the Granger causality test; we notice that there is

In a general way, the calculation of the average marginal effects (Tables 2, 4 and 7) demonstrates that the rate of children not attending school could decrease by 1.7%,

rural development in sub-Saharan Africa and Madagas- car), financed by the French Development Agency (AFD, www.afd.fr/en) and conducted in 2015-2016 by CIRAD in

Controls are survey month dummies, the number of adults and the number of children in the household, age dummies, the gender and the birth order of the child, the number of boys

Agriculture in the broad sense (including live- stock, fisheries and forestry) is a strategic economic sector for developing activities and employment, whether farm or

The demographic transition theory postulates that generally as a result of socio-economic development, western societies evolved from populations with high mortality and